Chronic sleep restriction and altered sleep patterns have emerged as a common and serious health and safety issue, and there is growing evidence that disrupted sleep directly underlies and/or contributes to metabolic abnormalities and diseases, including diabetes. Despite the obvious need, however, there are as yet no objective means of assessing chronic sleep status and limited understanding of intermediate phenotypes, especially in diverse populations. Not surprisingly, the first goal of the 2011 NIH Sleep Disorders Research Plan is to identify: ...metabolic... biomarkers of sleep deficiency ... that will facilitate personalized treatments, and clarify the risk associated with untreated sleep and circadian disorders and disturbances. It is biologically plausible that the metabolic consequences of restricted/altered sleep are, at least at some level, reflected in shifts within an individual's circulating lipids and overall adiposity status -- i.e., the clinical hallmarks that associate with, precede, and predict, metabolic syndrome and overt diabetes. Conversely, healthy, physically active adults display a fat distribution with a relatively low level of visceral and liver adiposity ? regardless of their overall adiposity. The heterogeneity in level and types of obesity that exists among the five ethnic groups in the Multiethnic Cohort (MEC) offers a unique research setting to better understand the ill effects of abnormalities in sleep duration and patterns, how these changes relate to diabetes risk, and how this relationship is modified by age and ethnicity. More specifically, we propose a general viewpoint -- and set of hypotheses, in which: (i) Sleep restriction and/or altered sleep patterns directly induce altered circulating lipids short-term and altered adiposity phenotypes (e.g., shifts in the ratios between visceral, liver, and subcutaneous fat) long-term; (ii) the core risk of insufficient sleep or altered sleep patterns is ethnic and sex- independent, but its penetrance on diabetes risk is influenced by the underlying risk within the different ethnic groups; (iii) the consequences of sleep problems continue and increase risk of mortality in diabetics specifically and, more generally, in the MEC. This proposal leverages data on ~1850 older individuals (age~65-70) DEXA/MRI-assessed for body fat/distribution (e.g., visceral fat, liver fat, subcutaneous fat, etc) and the availability of 2400 nested (in MEC), ethnically- and age-diverse (age~50 to 80) case-control pairs for future diabetes, as well as existing clinical and sleep study data in Brigham and Women's Hospital. Identification of biomarker panels for insufficient or abnormal sleep will impact multiple aspects of science and health: (i) contribute to clinical recognition and treatment (e.g., counseling, pharmaceuticals) of sleep issues; (ii) create epidemiologic tools to relate limited sleep with disease risk and aid development of other disease biomarkers, and; (iii) contribute to research on sleep biology and its implications for human health. Completion will further NIH goals of focusing on health and early interventions rather than late stage disease. The study is well within our lab's capacity to complete in 4 years.